A clustering algorithm that will perform clustering on each of a time-series of discrete (not a data stream... yet) datasets, and explicitly track the evolution of clusters over time.
If you use the ChronoClust algorithm, please cite the associated publication:
ChronoClust: Density-based clustering and cluster tracking in high-dimensional time-series data. Givanna H. Putri, Mark N. Read, Irena Koprinska, Deeksha Singh, Uwe Rohm, Thomas M. Ashhurst, Nick J.C. King. Accepted to Knowledge Based Systems, 2019.
DOI: https://doi.org/10.1016/j.knosys.2019.02.018
To run the project you will require the following packages for python 3:
- pandas
- numpy
- scipy
- scikit-learn
- pickle
- tqdm
- deprecation
- Givanna Putri ghar1821@sydney.edu.au